SOTAVerified

Model Compression

Model Compression is an actively pursued area of research over the last few years with the goal of deploying state-of-the-art deep networks in low-power and resource limited devices without significant drop in accuracy. Parameter pruning, low-rank factorization and weight quantization are some of the proposed methods to compress the size of deep networks.

Source: KD-MRI: A knowledge distillation framework for image reconstruction and image restoration in MRI workflow

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 1–50 of 1356 papers

TitleStatusHype
SANA 1.5: Efficient Scaling of Training-Time and Inference-Time Compute in Linear Diffusion TransformerCode9
GPTQ: Accurate Post-Training Quantization for Generative Pre-trained TransformersCode7
A Survey on Knowledge Distillation of Large Language ModelsCode5
LLM Inference Unveiled: Survey and Roofline Model InsightsCode4
Compact 3D Gaussian Splatting for Static and Dynamic Radiance FieldsCode3
ZipNN: Lossless Compression for AI ModelsCode3
SVD-LLM V2: Optimizing Singular Value Truncation for Large Language Model CompressionCode3
SVD-LLM: Truncation-aware Singular Value Decomposition for Large Language Model CompressionCode3
Efficient Reasoning Models: A SurveyCode3
ABQ-LLM: Arbitrary-Bit Quantized Inference Acceleration for Large Language ModelsCode3
LightGNN: Simple Graph Neural Network for RecommendationCode2
On-Device Domain GeneralizationCode2
PromptMM: Multi-Modal Knowledge Distillation for Recommendation with Prompt-TuningCode2
Diffusion Models for Image Restoration and Enhancement -- A Comprehensive SurveyCode2
Compact 3D Gaussian Representation for Radiance FieldCode2
Well-Read Students Learn Better: On the Importance of Pre-training Compact ModelsCode2
OmniQuant: Omnidirectionally Calibrated Quantization for Large Language ModelsCode2
QuEST: Low-bit Diffusion Model Quantization via Efficient Selective FinetuningCode2
Compressing Volumetric Radiance Fields to 1 MBCode2
Q-DiT: Accurate Post-Training Quantization for Diffusion TransformersCode2
LiDAR-PTQ: Post-Training Quantization for Point Cloud 3D Object DetectionCode2
Torch2Chip: An End-to-end Customizable Deep Neural Network Compression and Deployment Toolkit for Prototype Hardware Accelerator DesignCode2
Learning Student Networks in the WildCode2
MaskLLM: Learnable Semi-Structured Sparsity for Large Language ModelsCode2
Fast convolutional neural networks on FPGAs with hls4mlCode2
AMC: AutoML for Model Compression and Acceleration on Mobile DevicesCode2
Data-Free Knowledge Distillation for Deep Neural NetworksCode2
MoA: Mixture of Sparse Attention for Automatic Large Language Model CompressionCode2
Knowledge Distillation and Student-Teacher Learning for Visual Intelligence: A Review and New OutlooksCode2
Towards Lightweight Super-Resolution with Dual Regression LearningCode2
Contrastive Representation DistillationCode1
3DG-STFM: 3D Geometric Guided Student-Teacher Feature MatchingCode1
CPrune: Compiler-Informed Model Pruning for Efficient Target-Aware DNN ExecutionCode1
Designing Large Foundation Models for Efficient Training and Inference: A SurveyCode1
Constraint-aware and Ranking-distilled Token Pruning for Efficient Transformer InferenceCode1
Contrastive Distillation on Intermediate Representations for Language Model CompressionCode1
CrossKD: Cross-Head Knowledge Distillation for Object DetectionCode1
CompRess: Self-Supervised Learning by Compressing RepresentationsCode1
Compression-Aware Video Super-ResolutionCode1
Computation-Efficient Knowledge Distillation via Uncertainty-Aware MixupCode1
Streamlining Redundant Layers to Compress Large Language ModelsCode1
AD-KD: Attribution-Driven Knowledge Distillation for Language Model CompressionCode1
Composable Interventions for Language ModelsCode1
Consistent Quantity-Quality Control across Scenes for Deployment-Aware Gaussian SplattingCode1
DarwinLM: Evolutionary Structured Pruning of Large Language ModelsCode1
Communication-Computation Trade-Off in Resource-Constrained Edge InferenceCode1
COMCAT: Towards Efficient Compression and Customization of Attention-Based Vision ModelsCode1
Compacting, Picking and Growing for Unforgetting Continual LearningCode1
Comprehensive Knowledge Distillation with Causal InterventionCode1
Communication-Efficient Diffusion Strategy for Performance Improvement of Federated Learning with Non-IID DataCode1
Show:102550
← PrevPage 1 of 28Next →
All datasetsImageNetQNLI

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1ADLIK-MO-ResNet50+W4A4Top-177.88—Unverified
2ADLIK-MO-ResNet50+W3A4Top-177.34—Unverified
3ResNet-18 + 4bit-1dim model compression using DKMTop-170.52—Unverified
4MobileNet-v1 + 4bit-1dim model compression using DKMTop-169.63—Unverified
5ResNet-18 + 2bit-1dim model compression using DKMTop-168.63—Unverified
6MobileNet-v1 + 2bit-1dim model compression using DKMTop-167.62—Unverified
7ResNet-18 + 4bit-4dim model compression using DKMTop-166.1—Unverified
8ResNet-18 + 2bit-2dim model compression using DKMTop-164.7—Unverified
9MobileNet-v1 + 4bit-4dim model compression using DKMTop-161.4—Unverified
10ResNet-18 + 1bit-1dim model compression using DKMTop-159.7—Unverified
#ModelMetricClaimedVerifiedStatus
1MobileBERT + 2bit-1dim model compression using DKMAccuracy82.13—Unverified
2MobileBERT + 1bit-1dim model compression using DKMAccuracy63.17—Unverified